CN1849648A - Coding apparatus and decoding apparatus - Google Patents
Coding apparatus and decoding apparatus Download PDFInfo
- Publication number
- CN1849648A CN1849648A CNA2004800259646A CN200480025964A CN1849648A CN 1849648 A CN1849648 A CN 1849648A CN A2004800259646 A CNA2004800259646 A CN A2004800259646A CN 200480025964 A CN200480025964 A CN 200480025964A CN 1849648 A CN1849648 A CN 1849648A
- Authority
- CN
- China
- Prior art keywords
- frequency spectrum
- frequency band
- frequency
- spectrum
- obtains
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001228 spectrum Methods 0.000 claims abstract description 294
- 238000000034 method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 abstract description 38
- 238000006243 chemical reaction Methods 0.000 description 36
- 238000010586 diagram Methods 0.000 description 34
- 238000005070 sampling Methods 0.000 description 21
- 238000012545 processing Methods 0.000 description 13
- 238000004891 communication Methods 0.000 description 11
- 230000005236 sound signal Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000003595 spectral effect Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000006378 damage Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/038—Speech enhancement, e.g. noise reduction or echo cancellation using band spreading techniques
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/02—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/02—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders
- G10L19/0204—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders using subband decomposition
- G10L19/0208—Subband vocoders
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/08—Determination or coding of the excitation function; Determination or coding of the long-term prediction parameters
- G10L19/09—Long term prediction, i.e. removing periodical redundancies, e.g. by using adaptive codebook or pitch predictor
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Closures For Containers (AREA)
Abstract
An encoder apparatus capable of encoding spectrum with a high quality at a low bit rate without occurrence of disturbance in the harmonic structure of the spectrum. In the apparatus, an internal status setting part (106) uses a first spectrum S1(k) to set an internal status of a filtering part (107). A pitch coefficient setting part (109) changes a little by a little and outputs the pitch coefficient T. The filtering part (107) calculates, based on the pitch coefficient T, an estimated value S'2(k) of a second spectrum S2(k). A searching part (108) calculates a similarity between S2(k) and S'2(k). Then, a pitch coefficient T' having the maximum similarity as calculated is applied to a filter coefficient calculating part (110), which uses the pitch coefficient T' to obtain a filter coefficient beta<SUB>i</SUB>.
Description
Technical field
The present invention relates to be installed on radio communication device etc. the code translator that is used for the code device that voice signal, sound signal etc. is encoded and is used for this coded signal is deciphered.
Background technology
In mobile communication system, on the viewpoint of the transmission path capacity (channel capacity) that uses radiowave etc. effectively and recording medium, the coding techniques that voice signal, sound signal etc. is compressed into the low bit rate signal is a particular importance.
The example that is used for voice coding scheme that voice signal is encoded comprises the schemes such as standardized G726, G729 by ITU-T (international telecommunication union telecommunication's Standardization Sector).These schemes are used narrow band signals, and (300Hz～3.4kHz) is for the coding target and can carry out high-quality coding with the bit rate of 8kbit/s～32kbit/s.But such narrow band signal is that so narrow so that frequency band is 3.4kHz to the maximum, and its quality gives the audience with the in a muffled voice unclear impression of sound, and causes lacking the problem of the sense of reality.
In addition, (50Hz～7kHz) is the voice coding scheme of coding target to use broadband signal in addition.This typical case is G722, the G722.1 of ITU-T and the AMR-WB of 3GPP (third generation collaborative project) etc.These schemes can be carried out the coding of wideband speech signal with the bit rate of 6.6kbit/s～64kbit/s.But, when the signal that will be encoded is voice, though broadband signal has higher quality, when with the sound signal being target or when requiring the voice signal of realistic higher quality, this remains not enough.
On the other hand,, just might obtain to be equivalent to the wireless sense of reality of FM,, just might obtain the quality that can compare favourably with CD (compact disk) when highest frequency during up to the magnitude of 20kHz when the highest frequency of signal usually during magnitude at 10～15kHz.For such signal, the audio coding of being represented by MPEG (M moves motion picture expert group version) standardized layer III scheme and AAC scheme is appropriate.But these audio coding schemes have wide signal band and will be encoded, the problem that it causes the bit rate of the signal that is encoded to increase.
The example of traditional coding techniques comprises the technology (for example, with reference to patent documentation 1) of wide band signal being encoded with low bit rate.According to this technology, input signal is divided into the signal in low frequency territory and the signal in high-frequency territory, the frequency spectrum of the signal in usefulness low frequency territory replaces the frequency spectrum of the signal in high-frequency territory, and it is encoded, and reduces whole bit rate thus.
Figure 1A～Fig. 1 D shows the general introduction with the above-mentioned processing of the frequency spectrum in the spectrum replacement high-frequency territory in low frequency territory.Though this processing is used for carrying out in conjunction with encoding process at first, for the purpose of simplifying the description, description is wherein carried out the situation of above-mentioned processing as example on original signal.
Figure 1A shows the frequency spectrum that frequency band is restricted to the original signal on 0≤k＜FH, Figure 1B show be restricted to the signal on 0≤k＜FL frequency spectrum (wherein, FL＜FH), Fig. 1 C shows and uses above-mentioned technology, replaces the frequency spectrum in high-frequency territory (high frequency band) and the frequency spectrum that obtains with the frequency spectrum of low frequency territory (low-frequency band), and Fig. 1 D shows by according to the spectrum envelope information of the frequency spectrum of having replaced about this, the frequency spectrum that this replacement frequency spectrum of sharpening obtains.On these figure, transverse axis is represented frequency, and the longitudinal axis is represented spectrum intensity.
In this technology, service band is that low-frequency spectra (Figure 1B) the expression frequency band of 0≤k＜FL is the frequency spectrum (Figure 1A) of original signal of 0≤k＜FH.More particularly, (FL≤k＜FH) is by low-frequency spectra (0≤k＜FL) replace for high frequency spectrum.As the result of this processing, obtained the frequency spectrum shown in Fig. 1 C.In addition, for the purpose of simplifying the description, be that example describes with the situation of relation with FL=FH/2.According to the information of the spectrum envelope of relevant original signal, be adjusted at the amplitude of the frequency spectrum in the high-frequency territory of frequency spectrum of Fig. 1 C, obtain the frequency spectrum shown in Fig. 1 D.This frequency spectrum is the frequency spectrum by estimating that original signal obtains.
Patent documentation 1: international patent application 2001-521648 number country open (the 15th page, Fig. 1, Fig. 2).
Summary of the invention
In general, well-known, the frequency spectrum such as voice signal and sound signal has the harmonic structure that wherein at every turn occurs spectrum peak on the integral multiple [tone (pitch) that each is predetermined] of certain frequency.This harmonic structure is the important information that keeps the quality of voice signal, sound signal etc., and if disturbance appearred in the harmonic structure, then the audience will feel the deterioration of quality.
Fig. 2 A and Fig. 2 B are the diagrams that is used to illustrate prior art problems.
Fig. 2 A is the frequency spectrum that the frequency spectrum by the analyzing audio signal obtains.As shown in the drawing, original signal has the harmonic structure that has interval T on frequency axis.On the other hand, Fig. 2 B shows according to the result of the frequency spectrum of the original signal of above-mentioned technique computes and the frequency spectrum that obtains.When comparing these two frequency spectrums, from the frequency spectrum shown in Fig. 2 B, find out, at the low-frequency spectra S1 that replaces the source with replace among the high frequency spectrum S2 of destination and kept harmonic structure, and the harmonic structure in the link field between low-frequency spectra S1 and high frequency spectrum S2 (frequency spectrum S3) is destroyed.
When the spectral conversion with this calculating becomes time signal and listens to this signal, have the problem that makes audiences feel that quality worsens owing to such harmonic structure disturbance, the disturbance of this harmonic structure results from the shape of not considering harmonic structure and the fact of carrying out replacement.
The objective of the invention is to be to provide a kind of can encode to frequency spectrum high-qualityly and do not produce the code device of disturbance and the code translator that can decipher this encoded signals on the harmonic structure of this frequency spectrum with low bit rate.
Code device of the present invention adopts such structure, and it comprises: obtain part, it obtains the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band; Calculating section, calculates the parameter of the similarity of indication between low-frequency band frequency spectrum that is obtained and the high frequency band frequency spectrum that obtained at its harmonic structure based on frequency spectrum; And coded portion, it is encoded to the parameter of the indication similarity calculated rather than the high frequency band frequency spectrum that is obtained.
Code translator of the present invention adopts such structure, and it comprises: frequency spectrum obtains part, and it obtains the frequency spectrum of low-frequency band from the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band; The parameter acquiring part, it obtains the parameter of the similarity between the frequency spectrum of indication frequency spectrum of low-frequency band and high frequency band; And decoding unit, it uses low-frequency band frequency spectrum and the parameter obtained, the frequency spectrum of low-frequency band and high frequency band is deciphered.
Coding method of the present invention comprises: obtaining step, obtain the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band; Calculation procedure, calculate the parameter of the similarity of indication between low-frequency band frequency spectrum that is obtained and the high frequency band frequency spectrum that obtained based on the harmonic structure of frequency spectrum, and coding step, to the parameter of the indication similarity calculated, rather than the high frequency band frequency spectrum that is obtained encoded.
Interpretation method of the present invention comprises: the frequency spectrum obtaining step, obtain the frequency spectrum of low-frequency band from the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band; The parameter acquiring step is obtained the parameter of the similarity of indication between low-frequency band frequency spectrum and high frequency band frequency spectrum; And the decoding step, use low-frequency band frequency spectrum and the parameter obtained, the frequency spectrum of low-frequency band and high frequency band is deciphered.
The present invention can carry out the coding of frequency spectrum high-qualityly and not have that the harmonic structure of this frequency spectrum is had any destruction with low bit rate.In addition, the present invention can also improve tonequality when this encoded signals is deciphered.
Description of drawings
Fig. 1 is the diagram of explanation with the summary of the conventional process of the frequency spectrum in the spectrum replacement high-frequency territory in low frequency territory;
Fig. 2 is the diagram that is used to illustrate prior art problems;
Fig. 3 is the block diagram that illustrates according to the primary structure of the wireless base station apparatus of embodiment 1;
Fig. 4 is the block diagram that illustrates according to the inner structure of the code device of embodiment 1;
Fig. 5 is the block diagram that illustrates according to the spectrum coding inner structure partly of embodiment 1;
Fig. 6 is the diagram of explanation according to the summary of the Filtering Processing of the filtering part of embodiment 1;
Fig. 7 is the diagram how frequency spectrum of the estimated value of explanation the 2nd frequency spectrum changes along with the variation of tone (pitch) coefficient T;
Fig. 8 is the diagram how frequency spectrum of the estimated value of explanation the 2nd frequency spectrum changes along with the variation of tone coefficient T;
Fig. 9 is the process flow diagram that the example of the series of algorithms that the processing that part carries out is set according to embodiment 1, by filtering part, search part, tone coefficient is shown;
Figure 10 is the block diagram that illustrates according to the primary structure of the radio system of embodiment 1;
Figure 11 is the block diagram that illustrates according to the inner structure of the code translator of embodiment 1;
Figure 12 is the block diagram that illustrates according to the frequency spectrum decoding inner structure partly of embodiment 1;
Figure 13 is the diagram of deciphering frequency spectrum that explanation partly generates according to embodiment 1, by filtering;
Figure 14 A be illustrate when the code device according to embodiment 1 is applied to wired communication system, at the block diagram of the primary structure of transmitting terminal;
Figure 14 B be illustrate when the code translator according to embodiment 1 is applied to wired communication system, at the block diagram of the primary structure of receiving end;
Figure 15 is the block diagram that illustrates according to the spectrum coding primary structure partly of embodiment 2;
Figure 16 is the diagram that explanation is carried out the general introduction of filtering according to embodiment 2, use wave filter;
Figure 17 is the block diagram that illustrates according to the spectrum coding primary structure partly of embodiment 3;
Figure 18 is the block diagram that illustrates according to the frequency spectrum decoding primary structure partly of embodiment 4; And
Figure 19 is the block diagram that illustrates according to the frequency spectrum decoding primary structure partly of embodiment 5.
Embodiment
The present inventor is conceived to the characteristic as voice signal, sound signal etc. (below be referred to as acoustic signal), that is to say, acoustic signal forms the fact of the characteristic of harmonic structure in the frequency axis direction, find to use the frequency spectrum of some frequency bands in the frequency spectrum of full range band to carry out the possibility of coding of the frequency spectrum of residue frequency band, thereby realized the present invention.
That is to say, essence of the present invention is, for example, when when the signal spectrum that is divided into two frequency bands in high-frequency territory and low frequency territory is encoded, for high-frequency territory frequency spectrum is determined similarity between the frequency spectrum in high-frequency territory and low frequency territory, and the parameter of indicating this similarity carried out coding.
Below with reference to the accompanying drawings embodiments of the present invention are elaborated.
(embodiment 1)
Fig. 3 shows when the wireless encoding device according to embodiments of the present invention 1 is installed in the transmitting terminal of wireless communication system, the block diagram of the primary structure of wireless base station apparatus 130.
This wireless base station apparatus 130 comprises code device 120, input media 131, A/D conversion equipment 132, RF modulating device 133 and antenna 134.
Fig. 4 is the block diagram that the inner structure of above-mentioned code device 120 is shown.At this, be that example describes with the situation of wherein carrying out hierarchical coding (scalable coding).
The signal that has effective band and be 0≤k＜FH is input to input end 121 from A/D conversion equipment 132.Down-sampling part 122 is to the signal application down-sampling via input end 121 inputs, and generation has the signal of low sampling rate and exports this signal.The 1st layer of encoding section 123 encoded to the signal of this down-sampling, and to the coding that multiplexing section (multiplexer) 127 outputs are obtained, also exports the coding that is obtained to the 1st layer of decoding part 124 simultaneously.The 1st layer of decoding part 124 generates the 1st layer decoded signal based on this coding.The sampling rate of the decoded signal of the 1st layer of coded portion 123 improves in up-sampling portion 125.
On the other hand, decay part 126 provides the delay of specified length to the signal via input end 121 inputs.The length of supposing this delay has identical value with the time delay that produces during by down-sampling part 122, the 1st layer of coded portion 123, the 1st layer of decoding part 124 and up-sampling part 125 when signal.Spectrum coding part 100 is used from the signal of up-sampling part 125 outputs and is carried out spectrum coding as first signal and from the signal of decay part 126 outputs as secondary signal, and the coding that generates to multiplexing section 127 outputs.Multiplexing section 127 will carry out multiplexed from the 1st layer of coded portion 123 coding that obtains and the coding that obtains from spectrum coding part 100, and export these multiplexed parameters as output encoder via output terminal 128.Give RF modulating device 133 with the coding of this output.
Fig. 5 is the block diagram that the inner structure of above-mentioned spectrum coding part 100 is shown.
Spectrum coding part 100 comprises input end 102,103, frequency field conversion portion 104,105, internal state are provided with part 106, filtering part 107, search part 108, tone coefficient part 109, filter coefficient calculating section 110, and output terminal 111 are set.
First signal is input to input end 102 from up-sampling part 125.This first signal is to use the signal of being deciphered by the coding parameter of the 1st layer of coded portion 123 codings by the 1st layer of decoding part 124, and has the effective band of 0≤k＜FL.In addition, effective band is that (secondary signal of FL＜FH) is input to input end 103 from decay part 126 to 0≤k＜FH.
Frequency field conversion portion 104 is carrying out frequency inverted from first signal of input end 102 inputs, to calculate the first frequency spectrum S1 (k).In addition, frequency field conversion portion 105 is carrying out frequency inverted from the secondary signal of input end 103 inputs, to calculate the second frequency spectrum S2 (k).Herein, frequency conversion method is used the discrete cosine transform (MDCT) of discrete Fourier transformation (DFT), discrete cosine transform (DCT), modification etc.
Internal state is provided with part 106, and to use effective bands be the internal state that the first frequency spectrum S1 (k) of 0≤k＜FL is arranged on the wave filter that uses in the filtering part 107.The back will describe this setting once more.
The tone coefficient is provided with part 109 the hunting zone Tmin of pre-tone coefficient T bit by bit changes this coefficient in Tmax when, exports this tone coefficient T one by one to filtering part 107.
Filtering part 107 is provided with the tone coefficient T of part 109 outputs, the filtering of execution second frequency spectrum based on the internal state of the wave filter that part 106 settings are set by internal state and from the tone coefficient, and calculates the estimated value S ' 2 (k) of second frequency spectrum.Will be described later the details of this Filtering Processing.
Filter coefficient calculating section 110 is provided by the tone coefficient T that provides from search part 108 ' calculating filter coefficient β
i, and via output terminal 111 these filter coefficients of output.At this moment, simultaneously also via output terminal 111 output tone coefficient T '.
Next, will use mathematic(al) representation to describe the concrete operations of the major part of spectrum coding portion 100 in detail below.
Fig. 6 has illustrated the general introduction of the Filtering Processing of filtering part 107.
At this, (frequency spectrum of 0≤k＜FH) is called " S (k) ", and will use by the represented filter function of following equation with the full range band in supposition for convenience.
In this equation, z represents the z converted variable, and T represents from the tone coefficient coefficient that part 109 provides to be set, and supposition M=1.
As shown in the drawing, as the internal state of wave filter, the first frequency spectrum S1 (k) is stored in 0≤k＜FL frequency band of S (k).On the other hand, the estimated value S ' 2 (k) of second frequency spectrum that will obtain according to following process is stored in FL≤k＜FH frequency band of S (k).
In whole Filtering Processing, in S ' 2 (k), replace by the represented frequency spectrum of following equation (2).By with all frequency spectrum β
iS (k-T-i) adds up mutually and obtains this replacement spectrum, wherein by will frequency spectrum S (k-T) than the frequency of the low T of k is the center, only the be separated by contiguous frequency spectrum S (k-T-i) of i multiply by predetermined weighting coefficient β to have
iAnd acquisition β
iS (k-T-i).
At this moment, suppose that the input signal that offers this wave filter is zero.That is zero input of (equation 2) expression (equation 1).By in the scope of FL≤k＜FH, carrying out aforementioned calculation in increasing order (k=FL) the change k with frequency, calculate the estimated value S ' 2 (k) of second frequency spectrum in FL≤k＜FH.
Be provided with when providing the tone coefficient T the part 109 from the tone coefficient at every turn, by at every turn all with S (k) zero clearing, in the scope of FL≤k＜FH, carry out above-mentioned Filtering Processing.That is, just calculate S (k) when each tone coefficient T changes, and it is outputed to search part 108.
Then, the calculation of similarity degree of being carried out by search part 108 is handled and the derivation of optimum pitch coefficient T is handled with describing.
At first, have the definition of various similarities.Herein, will be wherein to suppose filter coefficient β
-1And β
1Be 0, use by situation and describe as example based on the defined similarity of following equation of Minimum Variance method.
Under the situation of having used this similarity, after calculating the optimum pitch coefficient T, determine filter coefficient β
iHere, E represents the variance between S2 (k) and the S ' 2 (k).In this equation, the 1st of the right becomes and the irrelevant fixed value of tone coefficient T, and therefore search is used to generate the tone coefficient T that makes the 2nd on the right become maximum S ' 2 (k).The 2nd on the right of this equation is called as " similarity ".
Fig. 7 A～Fig. 7 E is the figure that has illustrated how the frequency spectrum of the estimated value S ' 2 (k) of second frequency spectrum changes along with the variation of tone coefficient T.
Fig. 7 A is the figure that first frequency spectrum with the harmonic structure that is stored as internal state has been described.In addition, to be explanation carry out the figure of frequency spectrum of the estimated value S ' 2 (k) of second frequency spectrum that filtering calculates by the tone coefficient T 0, T1, T2 of using three types to Fig. 7 B～Fig. 7 D.Fig. 7 E shows the figure of the second frequency spectrum S2 (k) that will compare with the frequency spectrum of estimated value S ' 2 (k).
In the example shown in this figure,, and therefore can understand the similarity of using T1 to calculate and represent the highest value because the frequency spectrum shown in Fig. 7 C is similar to the frequency spectrum shown in Fig. 7 E.That is, T1 is as the optimal value that can keep the tone coefficient T of harmonic structure.
Fig. 8 A～Fig. 8 E also is and Fig. 7 A～similar figure of Fig. 7 E, and still, that of phase place and Fig. 7 A～Fig. 7 E of first frequency spectrum that is stored as internal state is different.Yet in the example shown in this figure, the tone coefficient T that keeps harmonic structure whereby also is T1.
Therefore, change tone coefficient T and search maximum similarity T and be equal to the tone (perhaps its integral multiple) of on the basis of souning out, finding out the harmonic structure of this frequency spectrum.Code device in this embodiment is calculated the estimated value S ' 2 (k) of second frequency spectrum based on the tonometer of this harmonic structure, and therefore the harmonic structure in the coupling part between first frequency spectrum and estimated spectral does not have destroyed.Consider first frequency spectrum that separates based on the tone T (perhaps its integral multiple) by harmonic structure, the estimated value S ' 2 (k) that calculates coupling part when k=FL, this is readily appreciated that.
In addition, the tone coefficient T is represented the integral multiple (round values) of the frequency interval of frequency spectrum data.Yet the tone of actual harmonic structure often is a non integer value.Therefore, by selecting suitable weighting coefficient β
iAnd to being that M proximity data at center used weighting summation with T, the tone of the harmonic structure of the non integer value in just might being illustrated in from T-M to the T+M scope.
Fig. 9 shows the process flow diagram that the example of the series of algorithms in the processing that part 109 carries out is set by filtering part 107, search part 108, tone coefficient.Owing to the general introduction of these processing was described, so will omit the detailed description of this flow process.
Next, will the computing of the filter coefficient that is undertaken by filter coefficient calculating section 110 be described.
The tone coefficient T that provides from search part 108 is provided for filter coefficient calculating section 110 ', determine to be minimized in the filter coefficient β of square distortion E the following equation
i
Filter coefficient calculating section 110 keeps a plurality of β in advance
iThe combination of (i=-1,0,1) determines to minimize the β of square distortion E in above-mentioned (equation 4) as tables of data
iThe combination of (i=-1,0,1), and export its index.
Therefore, for being divided into low frequency territory (0≤k＜FL) and the high-frequency territory (frequency spectrum of the input signal of two parts of FL≤k＜FH), code device in this embodiment uses and comprises the shape of low-frequency spectra as the filtering part 107 estimation high frequency spectrums of internal state, parameter rather than high frequency spectrum to the filter characteristic of indication filtering part 107 are encoded, and export this parameter, therefore, might carry out the coding of frequency spectrum with low bit rate high-qualityly.
In addition, in said structure, when filtering part 107 uses low-frequency spectra to estimate the shape of high frequency spectrum, the tone coefficient is provided with part 109 and changes in every way at low-frequency spectra that is used as the estimation standard and the difference on the frequency between the high frequency spectrum, promptly, the tone coefficient T, and export this difference on the frequency, and also search part 108 detects and the corresponding T of the maximum similarity between low-frequency spectra and high frequency spectrum.Therefore, may estimate the shape of high frequency spectrum based on the tone of the harmonic structure of entire spectrum, and can when keeping the harmonic structure of entire spectrum, carry out coding.
The bandwidth of low-frequency spectra also need be set based on the tone of harmonic structure in addition.That is, do not need to make the bandwidth of low-frequency spectra and the tone of harmonic structure (perhaps its integral multiple) to be complementary, and bandwidth can at random be set.This is because said structure allows to be connected frequency spectrum at low-frequency spectra smoothly with coupling part between the high frequency spectrum, and the bandwidth of low-frequency spectra and the tone of harmonic structure are complementary.
Though present embodiment has illustrated the situation of M=1 in (equation 1) as example, M is not limited to this and can uses integer (natural number) more than 0 or 0.
In addition, though present embodiment has illustrated the code device of execution hierarchical coding (scalable coding) as example, above-mentioned spectrum coding part 100 can also be installed in based on other schemes and carry out on the code device of coding.
In addition, present embodiment has illustrated that spectrum coding part 100 wherein comprises the situation of frequency field conversion portion 104,105.These are necessary parts when the time domain signal is used as input signal, still directly frequency spectrum are input to therein in the pattern in the spectrum coding part 100, and the frequency field conversion portion is optional.
In addition, illustrated in present embodiment and wherein to have used low-frequency spectra, promptly used the standard of low-frequency spectra for coding, the situation that high frequency spectrum is encoded, but the method that is provided with as the frequency spectrum of standard is not limited thereto.But also may use high frequency spectrum that low-frequency spectra is carried out coding, and perhaps use the frequency spectrum of intermediate frequency band to carry out and encode for the standard of coding, to other regional frequency spectrums, though from effectively utilizing on the viewpoint of energy, they are not desired.
Figure 10 illustrates the block diagram of reception from the primary structure of the radio receiver 180 of the signal of wireless base station apparatus 130 transmissions.
This radio receiver 180 comprises antenna 181, RF rectification device 182, code translator 170, D/A conversion equipment 183 and output unit 184.
The digitally coded acoustic signal that antenna 181 is accepted as radiowave W12 is generated as the coding acoustic signal of the digital received of electric signal, and provides it to RF rectification device 182.RF rectification device 182 rectification are from the coding acoustic signal of the reception of antenna 181, the coding acoustic signal of generating solution modulation and provide it to code translator 170.
Figure 11 shows the block diagram of the inner structure of above-mentioned code translator 170.Here will be being that example describes to situation about being deciphered by the signal of hierarchical coding.
This code translator 170 comprises input end 171, separating part 172, the 1st layer of decoding part 173, up-sampling part 174, frequency spectrum decoding part 150 and output terminal 176,177.
The sample frequency of the 1st layer of decoded signal that provides from the 1st layer of decoding part 173 is provided for up-sampling part 174.The 1st layer of decoded signal of the up-sampling that will generate by separating part 172 separated coding with by up-sampling part 174 offers frequency spectrum decoding part 150.Frequency spectrum decoding part 150 is carried out the frequency spectrum decoding of describing after a while, generates to have the decoded signal of signal band 0≤k＜FH, and via output terminal 177 these decoded signals of output.Frequency spectrum decoding part 150 is regarded the 1st layer of decoded signal of the up-sampling that provides from up-sampling part 174 as the 1st signal and is carried out and handle.
According to this structure, when needs are exported the 1st layer of decoded signal that is generated by the 1st layer of decoding part 173, can be from the 1st layer of decoded signal of output terminal 176 outputs.In addition, when the higher output signal of the quality that needs output spectrum decoding part 150, this output signal can be from output terminal 177 outputs.Code translator 170 is exported the signal of any output from output terminal 176 or output terminal 177, and this signal is offered D/A conversion equipment 183.Export which signal and depend on the setting of application or user's judgement.
Figure 12 shows the block diagram of the inner structure of above-mentioned frequency spectrum decoding part 150.
This frequency spectrum decoding part 150 comprises that input end 152,153, frequency field conversion portion 154, internal state are provided with part 155, filtering part 156, time domain conversion portion 158 and output terminal 159.
To indicate the filter coefficient of the coding that obtains by spectrum coding part 100 to be input to input end 152 via separating part 172.In addition, the 1st signal that will have effective band 0≤k＜FL is input to input end 153.This 1st signal be by the decoding of the 1st layer of decoding part 173 and by 174 samplings of up-sampling part the 1st layer of decoded signal.
Frequency field conversion portion 154 is changed from the frequency of the time-domain signal of input end 153 inputs, and calculates the 1st frequency spectrum S1 (k).As frequency conversion method, can use the discrete cosine transform (MDCT) of discrete Fourier transformation (DFT), discrete cosine transform (DCT), modification etc.
Internal state is provided with part 155 and uses the 1st frequency spectrum S1 (k) to be arranged on the internal state of the wave filter that uses in the filtering part 156.
Filtering part 156 is based on wave filter internal state that part 155 is provided with and the tone coefficient T that provides from input end 152 are set by internal state ' and filter factor β carry out the filtering of the 1st frequency spectrum, and calculate the estimated value S ' 2 (k) of the 2nd frequency spectrum.In this case, filtering part 156 is used the filter function described in (equation 1).
The decoding frequency spectrum S ' that time domain conversion portion 158 will obtain from filtering part 156 (k) converts time-domain signal to, and via this decoding frequency spectrum of output terminal 159 outputs.At this, that the processing of carrying out as required such as the suitable addition of windowing and superpose etc. is discontinuous to avoid occurring between frame.
Figure 13 shows the decoding frequency spectrum S ' that generated by filtering part 156 (k).
As shown in the drawing, the decoding frequency spectrum S ' with frequency band 0≤k＜FL (k) is made of the 1st frequency spectrum S1 (k), and the decoding frequency spectrum S ' with frequency band FL≤k＜FH (k) is made of the estimated value S ' 2 (k) of the 2nd frequency spectrum.
Therefore, the code translator in the present embodiment has and according to the corresponding structure of the coding method of present embodiment, therefore can enough less figure places efficiently the acoustic signal of coding be deciphered, and can export high-quality acoustic signal.
Though at this has been that example is illustrated with the situation that wherein is applied to wireless communication system according to the code device or the code translator of present embodiment, but as follows, can be applicable to wired communication system equally according to the code device or the code translator of present embodiment.
Figure 14 A shows when the code translator according to present embodiment is applied to wired communication system, the block diagram of the primary structure of transmitting terminal.To the parts distribution identical reference number same, and will omit its explanation with parts shown in Figure 3.
Wired dispensing device 140 comprises code device 120, input media 131 and A/D conversion equipment 132, and its output is connected to network N 1.
The input end of A/D conversion equipment 132 is connected to the output terminal of input media 131.The input end of code device 120 is connected to the output terminal of A/D conversion equipment 132.The output terminal of code device 120 is connected to network N 1.
Figure 14 B shows when the code translator according to present embodiment is applied to wired communication system, the block diagram of the primary structure of receiving end.In addition, distribute identical reference number to the parts identical, and will omit its explanation with parts shown in Figure 10.
Wired receiving trap 190 comprises receiving trap 191, code translator 170, D/A conversion equipment 183 and the output unit 184 that is connected to network N 1.
The input end of receiving trap 191 is connected to network N 1.The input end of code translator 170 is connected to the output terminal of receiving trap 191.The input end of D/A conversion equipment 183 is connected to the output terminal of code translator 170.The input end of output unit 184 is connected to the output terminal of D/A conversion equipment 183.
Receiving trap 191 generates the acoustic signal of digital received from the receiving digitally encoded acoustic signal of network N 1, and this signal is offered code translator 170.Code translator 170 receives the acoustic signal that is received from receiving trap 191, carries out decoding and handle on the acoustic signal that is received, and generates the decoding acoustic signal of numeral, and provides it to D/A conversion equipment 183.183 conversions of D/A conversion equipment generate the decoding voice signal of simulation, and provide it to output unit 184 from the decoding voice signal of the numeral of code translator 170.Output unit 184 will convert the vibration of air as the decoding acoustic signal of the simulation of electric signal to, and with the audible sound wave W13 output of its ear as the people.
Therefore, according to said structure, can provide the wired R-T unit that has with above-mentioned wireless transmitter similar operation and effect.
(embodiment 2)
Figure 15 shows the block diagram of the primary structure of the spectrum coding part 200 in the code device of foundation embodiments of the present invention 2.This spectrum coding part 200 has the basic structure similar to spectrum coding part shown in Figure 5 100, and distributes identical reference number to identical parts, and will omit its explanation.
The characteristics of present embodiment are, make the filter function that uses in the filtering part simpler than in the embodiment 1 that.
For the filter function that uses in the filtering part 201, used simplification shown in following equation one.
This equation is corresponding to supposition M=0, β in (equation 1)
0=1 filter function.
Figure 16 has illustrated the general introduction of having used above-mentioned wave filter to carry out filtering.
Obtain the estimated value S ' 2 (k) of the 2nd frequency spectrum by the low-frequency spectra that sequentially duplicates the T of being separated by.In addition, as the situation at embodiment 1, search part 108 is determined the optimum pitch coefficient T by the tone coefficient T that search minimizes the E in (equation 3) '.Export the tone coefficient T that obtains by this way via output terminal 111 '.In this structure, the characteristic of wave filter is only determined by the tone coefficient T.
It is also noted that, the wave filter of present embodiment is characterised in that, it is operated in the mode that is similar to adaptive code book (adaptive codebook), wherein the adaptive code book is one of key element of CELP (Code-Excited Linear Prediction, the code-excited linear prediction) scheme as the representative art of low rate voice coding.
Next, with the frequency spectrum decoding part (not shown) of explanation to deciphering by above-mentioned spectrum coding part 200 encoded signals.
This frequency spectrum decoding part has the structure similar to frequency spectrum shown in Figure 12 decoding part 150, therefore omitted it and described in detail, and it has following characteristics.That is, when filtering part 156 was calculated the estimated value S ' 2 (k) of the 2nd frequency spectrum, it used the filter function described in (equation 5), rather than the filter function described in (equation 1).Tone coefficient T only ' provide from input end 152.That is, should use which the described filter function in (equation 1) or (equation 5), depend on the type of the filter function that uses at coding side and determine, and the use filter function identical with coding side.
Therefore, according to present embodiment, make the filter function that uses in the filtering part become simpler, it has caused eliminating necessity that the filter coefficient calculating section is installed.Therefore can enough less calculated amount estimate the 2nd frequency spectrum (high frequency spectrum), but also reduce circuit scale.
(embodiment 3)
Figure 17 shows the block diagram according to the primary structure of the spectrum coding part 300 in the code device of embodiments of the present invention 3.This spectrum coding part 300 has the basic structure similar to spectrum coding part shown in Figure 5 100, and has distributed identical reference number and will omit its explanation to identical parts.
The characteristics of present embodiment are, also comprise profile calculating section 301 and multiplexing section 302, and carry out after having estimated the 2nd frequency spectrum, to the envelope information of relevant the 2nd frequency spectrum and encode.
The estimated value S ' 2 (k) of search part 108 the 2nd frequency spectrum of ' output to multiplexing section 302, and will use this tone coefficient T ' generation with the optimum pitch coefficient T outputs to profile calculating section 301.Profile calculating section 301 is provided based on the 2nd frequency spectrum S2 (k) that provides from frequency field conversion portion 105 by the envelope information of relevant the 2nd frequency spectrum S2 (k).Here, wherein to represent that by the spectrum power that is used for each sub-band the situation that this envelope information and frequency band FL≤k＜FH are divided into J sub-band is that example describes.At this moment, the spectrum power of j sub-band is represented with following equation.
In this equation, the low-limit frequency of j sub-band of BL (j) expression, the highest frequency of j sub-band of BH (j) expression.The sub-band information of the 2nd frequency spectrum that obtains is by this way seen as the spectrum envelope information of relevant the 2nd frequency spectrum.
In a similar fashion, according to following formula calculate the 2nd frequency spectrum estimated value S ' 2 (k) sub-band information B ' (j).
And calculate the variable quantity V (j) of each sub-band according to following formula.
Next, 301 couples of variable quantity V of profile calculating section (j) encode, the variable quantity V after obtaining to encode
q(j), and with its index output to multiplexing section 302.The multiplexed optimum pitch coefficient T that obtain from search part 108 of multiplexing section 302 ' and from the index of the variable quantity V (j) of profile calculating section 301 outputs, and via output terminal 111 these multiplexed results of output.
Therefore because after having estimated high frequency spectrum a nearly step envelope information of relevant high frequency spectrum is encoded, become possibility so this embodiment make to improve the precision of the estimated value of high frequency spectrum.
(embodiment 4)
Figure 18 shows the block diagram according to the primary structure of the frequency spectrum decoding part 550 of embodiment of the present invention 4.This frequency spectrum decoding part 550 has the basic structure similar to frequency spectrum shown in Figure 12 decoding part 150, and has therefore distributed identical reference number and will omit its explanation to identical parts.
The characteristics of present embodiment are, also comprise separating part 551, spectrum envelope decoding part 552 and frequency spectrum adjustment member 553.Therefore, this allows spectrum coding part 300 grades shown in embodiment 3 to by envelope information being encoded and the estimated spectral of the high frequency spectrum coding that produces of encoding being carried out decoding.
Separating part 551 separates the coding via input end 152 inputs, and the information of relevant filter factor is offered filtering part 156, and the information of relevant spectrum envelope is offered spectrum envelope deciphers part 552.
Spectrum envelope decoding part 552 is deciphered out the variable quantity V that encodes and obtain by to variable quantity V (j) from the spectrum envelope information that is provided by separating part 551
q(j).
The decoding frequency spectrum S ' that frequency spectrum adjustment part 553 will obtain from filtering part 156 according to following equation (k) multiply by the variable quantity V of the decoding of each sub-band that obtains from spectrum envelope decoding part 552
q(j), be adjusted at the spectral shape among decoding frequency spectrum S ' frequency band FL≤k＜FH (k) and generate adjusted decoding frequency spectrum S3 (k).Should output to time domain conversion portion 158 by adjusted decoding frequency spectrum S3 (k), and convert it to time-domain signal.
S3 (k)=S ' is V (k)
q(j) (BL (j)≤k≤BH (j), for all j) ... (equation 9)
Therefore, according to present embodiment, might decipher the coding that comprises envelope information.
Present embodiment illustrated as spectrum envelope information example, that wherein provide from separating part 511 be by to shown in (equation 8), the variable quantity V (j) of each sub-band value V that encodes and obtain
q(j) situation, still, spectrum envelope information is not limited to this.
(embodiment 5)
Figure 19 is the block diagram that is illustrated in according to the primary structure of the decoding of the frequency spectrum in the code translator of embodiment of the present invention 5 part 650.This frequency spectrum decoding part 650 has the basic structure similar to frequency spectrum shown in Figure 180 decoding part 550, and therefore distributes identical reference number and will omit its explanation to identical parts.
The characteristics of present embodiment are, also comprise LPC frequency spectrum calculating section 652, utilize LPC frequency spectrum that the LPC coefficient calculations goes out as spectrum envelope information with use, estimate the 2nd frequency spectrum, then by the 2nd frequency spectrum and LPC frequency spectrum being multiply by mutually the acquisition estimated value of the 2nd frequency spectrum more accurately.
LPC frequency spectrum calculating section 652 calculates LPC frequency spectrum env (k) according to following equation from the LPC factor alpha (j) via input end 651 inputs.
Wherein, NP represents the number of times of LPC coefficient.In addition, also might use the characteristic of variable r (0＜r＜1) and change LPC frequency spectrum to calculate LPC frequency spectrum env (k).
In this case, LPC frequency spectrum env (k) is by shown in the following equation.
Wherein, r can be defined as fixed value, it also can adopt the value that changes frame by frame.The LPC frequency spectrum env (k) that calculates is by this way outputed to frequency spectrum adjustment member 553.
Frequency spectrum adjustment member 553 is according to following equation (12), to (k) multiply by the LPC frequency spectrum env (k) that obtains from LPC frequency spectrum calculating section 652 from the decoding frequency spectrum S ' that filtering part 156 obtains according to following formula, be adjusted at the frequency spectrum among decoding frequency spectrum S ' frequency band FL≤k＜FH (k), and generate adjusted decoding frequency spectrum S3 (k).Should offer time domain conversion portion 158 by adjusted decoding frequency spectrum S3 (k), and convert thereof into time-domain signal.
S3 (k)=S ' is env (k) (FL≤k＜FH) (k) ... (equation 12)
Therefore, according to present embodiment, use the LPC frequency spectrum to make and obtain more accurately that the estimated value of the 2nd frequency spectrum becomes possibility as spectrum envelope information.
Can be loaded on the communication terminal and base station apparatus in the mobile communication system according to code device of the present invention or code translator, and therefore can provide communication terminal and the base station apparatus that has to those above-mentioned similar operations and effect.
Though situation about being made of hardware with the present invention is that example is illustrated up to now,, the present invention also can realize with software.
The application is based on the Jap.P. of submitting on September 16th, 2003 2003-323658 number.The content of this application is incorporated at this by causing all.
Industrial Applicability A
Have the effect of carrying out coding with low bit rate according to code device of the present invention and code translator, and And also can be applied to wireless communication system etc.
Claims (14)
1, a kind of code device comprises:
Obtain part, obtain the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band;
Calculating section based on the harmonic structure of described frequency spectrum, calculates the parameter of the similarity of indication between described low-frequency band frequency spectrum that obtains and the described high frequency band frequency spectrum that obtains; And
Coded portion is encoded to described parameter calculated, the indication similarity, rather than the described high frequency band frequency spectrum that obtains is encoded.
2, code device as claimed in claim 1, wherein, described calculating section comprises;
Generating portion, based on frequency axis with the described low-frequency band frequency spectrum of described high frequency band frequency spectrum from the integral multiple of the tone of harmonic structure, generate the similar frequency spectrum of described high frequency band frequency spectrum; And
The parameter of the characteristic of the described similar frequency spectrum of indication when described similar frequency spectrum is the most similar to described high frequency band frequency spectrum, is detected in the test section.
3, code device as claimed in claim 1, wherein, described coded portion is also encoded to the information of the envelope of the relevant described high frequency band frequency spectrum that obtains.
4, code device as claimed in claim 1, wherein, described calculating section comprises:
Estimating part is used the wave filter of the described low-frequency band frequency spectrum that obtains as internal state, estimates the described high frequency band frequency spectrum that obtains; And
Output, the parameter of the characteristic of the described wave filter of output indication,
Wherein, the filter function of described wave filter represented by following equation,
And described estimating part uses the zero input response of described wave filter to carry out described estimation, in described equation,
P (z): filter function
The z:z transformed variable
M: multiply by 1/2 times wave filter number of times
β: weighting coefficient
T: tone coefficient.
5, code device as claimed in claim 4, wherein, in described filter function, M=0 and β
0=1.
6, code device as claimed in claim 1 wherein, obtains the frequency spectrum of described low-frequency band from the afterwards decoded signal that is encoded by hierarchical coding lower level.
7, a kind of R-T unit comprises code device as claimed in claim 1.
8, a kind of code translator comprises:
Frequency is obtained part, obtains the frequency spectrum of low-frequency band from the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band;
The parameter acquiring part is obtained the parameter of the similarity of indication between the frequency spectrum of the frequency spectrum of described low-frequency band and described high frequency band; And
The decoding part is used described low-frequency band frequency spectrum that obtains and described parameter, and the frequency spectrum of described low-frequency band and high frequency band is deciphered.
9, code translator as claimed in claim 8 comprises that also envelope information obtains part, obtains the information of the envelope of relevant described high frequency band frequency spectrum, and wherein, described decoding part is also used the described decoding of information and executing of the relevant described envelope that obtains.
10, the code translator shown in claim 8, wherein, described decoding part branch comprises estimating part, it comprises the described low-frequency band frequency spectrum that obtains as internal state, and use and have described getting parms as the wave filter of filter characteristic, the frequency spectrum of the described high frequency band of estimation, wherein, the filter function of above-mentioned wave filter is represented by following equation
And described estimating part uses the zero input response of described wave filter to carry out described estimation, in described equation,
P (z): filter function
The z:z transformed variable
M: multiply by 1/2 times wave filter number of times
β: weighting coefficient
T: tone coefficient.
11, code translator as claimed in claim 10, wherein, in described filter function, M=0 and β
0=1.
12, code translator as claimed in claim 8, wherein, the frequency spectrum of described low-frequency band generates from the signal of deciphering lower level by hierarchical coding.
13, a kind of R-T unit comprises code translator as claimed in claim 8.
14, a kind of coding method comprises:
Obtaining step obtains the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band;
Calculation procedure based on the harmonic structure of described frequency spectrum, is calculated the parameter of the similarity of indication between described low-frequency band frequency spectrum that obtains and the described high frequency band frequency spectrum that obtains; And
Coding step is encoded to parameter described calculating, the indication similarity, rather than the described high frequency band frequency spectrum that obtains is encoded.
15, a kind of interpretation method comprises:
The frequency spectrum obtaining step obtains the frequency spectrum of low-frequency band from the frequency spectrum that is divided into low-frequency band and two frequency bands of high frequency band;
The parameter acquiring step is obtained the parameter of the similarity of indication between described low-frequency band frequency spectrum that obtains and the described high frequency band frequency spectrum that obtains; And
The decoding step is used described low-frequency band frequency spectrum that obtains and described parameter, and the frequency spectrum of described low-frequency band and high frequency band is deciphered.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2003323658 | 2003-09-16 | ||
| JP323658/2003 | 2003-09-16 |
Related Child Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102658908A Division CN101800049B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
| CN2009102658895A Division CN101783141B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1849648A true CN1849648A (en) | 2006-10-18 |
| CN100590712C CN100590712C (en) | 2010-02-17 |
Family
ID=34308695
Family Applications (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102658908A Active CN101800049B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
| CN2009102658895A Active CN101783141B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
| CN200480025964A Active CN100590712C (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
Family Applications Before (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102658908A Active CN101800049B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
| CN2009102658895A Active CN101783141B (en) | 2003-09-16 | 2004-09-15 | Coding apparatus and decoding apparatus |
Country Status (6)
| Country | Link |
|---|---|
| EP (3) | EP2264700A1 (en) |
| JP (2) | JP4789622B2 (en) |
| CN (3) | CN101800049B (en) |
| BR (1) | BRPI0414444B1 (en) |
| DE (2) | DE602004021266D1 (en) |
| WO (1) | WO2005027095A1 (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009121298A1 (en) * | 2008-04-03 | 2009-10-08 | 华为技术有限公司 | A coding method, decoding method, system and device |
| WO2010022661A1 (en) * | 2008-08-28 | 2010-03-04 | 华为技术有限公司 | Method, apparatus and system for audio encoding and decoding |
| CN101436407B (en) * | 2008-12-22 | 2011-08-24 | 西安电子科技大学 | Method for encoding and decoding audio |
| CN103155031A (en) * | 2010-10-15 | 2013-06-12 | 索尼公司 | Encoding device and method, decoding device and method, and program |
| CN102708873B (en) * | 2007-01-12 | 2015-08-05 | 三星电子株式会社 | For the method for bandwidth extension encoding and decoding, equipment and medium |
| CN105408957A (en) * | 2013-06-11 | 2016-03-16 | 松下电器（美国）知识产权公司 | Device and method for bandwidth extension for acoustic signals |
| CN106663442A (en) * | 2014-07-28 | 2017-05-10 | 弗劳恩霍夫应用研究促进协会 | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2889526B1 (en) | 2005-08-04 | 2012-02-17 | Aventis Pharma Sa | SUBSTITUTED 7-AZA-INDAZOLES, COMPOSITIONS CONTAINING SAME, PROCESS FOR PRODUCTION AND USE |
| WO2007026763A1 (en) * | 2005-08-31 | 2007-03-08 | Matsushita Electric Industrial Co., Ltd. | Stereo encoding device, stereo decoding device, and stereo encoding method |
| US9159333B2 (en) | 2006-06-21 | 2015-10-13 | Samsung Electronics Co., Ltd. | Method and apparatus for adaptively encoding and decoding high frequency band |
| KR101390188B1 (en) * | 2006-06-21 | 2014-04-30 | 삼성전자주식회사 | Method and apparatus for encoding and decoding adaptive high frequency band |
| JP5294713B2 (en) * | 2007-03-02 | 2013-09-18 | パナソニック株式会社 | Encoding device, decoding device and methods thereof |
| JP4708446B2 (en) | 2007-03-02 | 2011-06-22 | パナソニック株式会社 | Encoding device, decoding device and methods thereof |
| US20100280833A1 (en) * | 2007-12-27 | 2010-11-04 | Panasonic Corporation | Encoding device, decoding device, and method thereof |
| CA2989886C (en) | 2008-12-15 | 2020-05-05 | Frederik Nagel | Audio encoder and bandwidth extension decoder |
| EP3598446B1 (en) | 2009-01-16 | 2021-12-22 | Dolby International AB | Cross product enhanced harmonic transposition |
| CN103718240B (en) | 2011-09-09 | 2017-02-15 | 松下电器（美国）知识产权公司 | Encoding device, decoding device, encoding method and decoding method |
| CN104517611B (en) | 2013-09-26 | 2016-05-25 | 华为技术有限公司 | A kind of high-frequency excitation signal Forecasting Methodology and device |
| EP3723086A1 (en) * | 2014-07-25 | 2020-10-14 | FRAUNHOFER-GESELLSCHAFT zur Förderung der angewandten Forschung e.V. | Audio signal coding apparatus, audio signal decoding apparatus, audio signal coding method, and audio signal decoding method |
| ES2707337T3 (en) * | 2014-07-25 | 2019-04-03 | Fraunhofer Ges Forschung | Audio signal encoding device, audio signal decoding device and methods thereof |
Family Cites Families (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0685607A (en) * | 1992-08-31 | 1994-03-25 | Alpine Electron Inc | High band component restoring device |
| JP3483958B2 (en) * | 1994-10-28 | 2004-01-06 | 三菱電機株式会社 | Broadband audio restoration apparatus, wideband audio restoration method, audio transmission system, and audio transmission method |
| JP3283413B2 (en) * | 1995-11-30 | 2002-05-20 | 株式会社日立製作所 | Encoding / decoding method, encoding device and decoding device |
| JP3243174B2 (en) * | 1996-03-21 | 2002-01-07 | 株式会社日立国際電気 | Frequency band extension circuit for narrow band audio signal |
| SE512719C2 (en) * | 1997-06-10 | 2000-05-02 | Lars Gustaf Liljeryd | A method and apparatus for reducing data flow based on harmonic bandwidth expansion |
| JP3539165B2 (en) * | 1997-11-12 | 2004-07-07 | 日本ビクター株式会社 | Code information processing method and apparatus, code information recording method on recording medium |
| CA2252170A1 (en) * | 1998-10-27 | 2000-04-27 | Bruno Bessette | A method and device for high quality coding of wideband speech and audio signals |
| SE9903553D0 (en) * | 1999-01-27 | 1999-10-01 | Lars Liljeryd | Enhancing conceptual performance of SBR and related coding methods by adaptive noise addition (ANA) and noise substitution limiting (NSL) |
| US6978236B1 (en) * | 1999-10-01 | 2005-12-20 | Coding Technologies Ab | Efficient spectral envelope coding using variable time/frequency resolution and time/frequency switching |
| JP3881836B2 (en) * | 2000-10-24 | 2007-02-14 | 株式会社ケンウッド | Frequency interpolation device, frequency interpolation method, and recording medium |
| SE0004163D0 (en) * | 2000-11-14 | 2000-11-14 | Coding Technologies Sweden Ab | Enhancing perceptual performance or high frequency reconstruction coding methods by adaptive filtering |
| JP3887531B2 (en) * | 2000-12-07 | 2007-02-28 | 株式会社ケンウッド | Signal interpolation device, signal interpolation method and recording medium |
| SE0101175D0 (en) * | 2001-04-02 | 2001-04-02 | Coding Technologies Sweden Ab | Aliasing reduction using complex-exponential-modulated filter banks |
| JP2003108197A (en) * | 2001-07-13 | 2003-04-11 | Matsushita Electric Ind Co Ltd | Audio signal decoding device and audio signal encoding device |
| EP1351401B1 (en) * | 2001-07-13 | 2009-01-14 | Panasonic Corporation | Audio signal decoding device and audio signal encoding device |
| JP2003044098A (en) * | 2001-07-26 | 2003-02-14 | Nec Corp | Device and method for expanding voice band |
| CN100395817C (en) * | 2001-11-14 | 2008-06-18 | 松下电器产业株式会社 | Encoding device and decoding device |
| JP3926726B2 (en) * | 2001-11-14 | 2007-06-06 | 松下電器産業株式会社 | Encoding device and decoding device |
| JP2003323658A (en) | 2002-05-07 | 2003-11-14 | Fujitsu Ltd | Paper money housing device with movable wall, method for controlling movable wall, and paper money depositor /dispenser |
-
2004
- 2004-09-15 CN CN2009102658908A patent/CN101800049B/en active Active
- 2004-09-15 CN CN2009102658895A patent/CN101783141B/en active Active
- 2004-09-15 BR BRPI0414444A patent/BRPI0414444B1/en active IP Right Grant
- 2004-09-15 EP EP10182526A patent/EP2264700A1/en not_active Withdrawn
- 2004-09-15 DE DE602004021266T patent/DE602004021266D1/en active Active
- 2004-09-15 JP JP2005513947A patent/JP4789622B2/en active Active
- 2004-09-15 WO PCT/JP2004/013455 patent/WO2005027095A1/en active Application Filing
- 2004-09-15 DE DE602004032587T patent/DE602004032587D1/en active Active
- 2004-09-15 EP EP04787895A patent/EP1657710B1/en active Active
- 2004-09-15 EP EP09156970A patent/EP2071565B1/en active Active
- 2004-09-15 CN CN200480025964A patent/CN100590712C/en active Active
-
2010
- 2010-10-18 JP JP2010233659A patent/JP5171922B2/en active Active
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708873B (en) * | 2007-01-12 | 2015-08-05 | 三星电子株式会社 | For the method for bandwidth extension encoding and decoding, equipment and medium |
| WO2009121298A1 (en) * | 2008-04-03 | 2009-10-08 | 华为技术有限公司 | A coding method, decoding method, system and device |
| WO2010022661A1 (en) * | 2008-08-28 | 2010-03-04 | 华为技术有限公司 | Method, apparatus and system for audio encoding and decoding |
| CN101436407B (en) * | 2008-12-22 | 2011-08-24 | 西安电子科技大学 | Method for encoding and decoding audio |
| CN103155031A (en) * | 2010-10-15 | 2013-06-12 | 索尼公司 | Encoding device and method, decoding device and method, and program |
| CN103155031B (en) * | 2010-10-15 | 2015-04-01 | 索尼公司 | Encoding device and method, decoding device and method |
| CN105408957A (en) * | 2013-06-11 | 2016-03-16 | 松下电器（美国）知识产权公司 | Device and method for bandwidth extension for acoustic signals |
| US10157622B2 (en) | 2013-06-11 | 2018-12-18 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Device and method for bandwidth extension for audio signals |
| US10522161B2 (en) | 2013-06-11 | 2019-12-31 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Device and method for bandwidth extension for audio signals |
| CN105408957B (en) * | 2013-06-11 | 2020-02-21 | 弗朗霍弗应用研究促进协会 | Apparatus and method for band extension of voice signal |
| CN106663442A (en) * | 2014-07-28 | 2017-05-10 | 弗劳恩霍夫应用研究促进协会 | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| US11170797B2 (en) | 2014-07-28 | 2021-11-09 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| US11922961B2 (en) | 2014-07-28 | 2024-03-05 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101800049B (en) | 2012-05-23 |
| CN101783141B (en) | 2012-06-20 |
| EP2071565A3 (en) | 2009-07-08 |
| CN100590712C (en) | 2010-02-17 |
| EP1657710A4 (en) | 2007-10-31 |
| CN101783141A (en) | 2010-07-21 |
| WO2005027095A1 (en) | 2005-03-24 |
| EP1657710A1 (en) | 2006-05-17 |
| EP2264700A1 (en) | 2010-12-22 |
| JP5171922B2 (en) | 2013-03-27 |
| BRPI0414444B1 (en) | 2020-05-05 |
| JP4789622B2 (en) | 2011-10-12 |
| CN101800049A (en) | 2010-08-11 |
| JP2011039553A (en) | 2011-02-24 |
| DE602004032587D1 (en) | 2011-06-16 |
| JPWO2005027095A1 (en) | 2007-11-08 |
| EP2071565B1 (en) | 2011-05-04 |
| EP1657710B1 (en) | 2009-05-27 |
| BRPI0414444A (en) | 2006-11-14 |
| DE602004021266D1 (en) | 2009-07-09 |
| EP2071565A2 (en) | 2009-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1096148C (en) | Signal encoding method and apparatus | |
| CN1849648A (en) | Coding apparatus and decoding apparatus | |
| CN1308916C (en) | Source coding enhancement using spectral-band replication | |
| CN1689069A (en) | Sound encoding apparatus and sound encoding method | |
| CN100338649C (en) | Reconstruction of the spectrum of an audiosignal with incomplete spectrum based on frequency translation | |
| CN1781141A (en) | Improved audio coding systems and methods using spectral component coupling and spectral component regeneration | |
| CN1910655A (en) | Apparatus and method for constructing a multi-channel output signal or for generating a downmix signal | |
| CN1795495A (en) | Audio encoding device, audio decoding device, audio encodingmethod, and audio decoding method | |
| CN101048814A (en) | Encoder, decoder, encoding method, and decoding method | |
| CN1202514C (en) | Method, device and program for coding and decoding acoustic parameter, and method, device and program for coding and decoding sound | |
| CN101053019A (en) | Encoding and decoding of audio signals using complex-valued filter banks | |
| CN1161751C (en) | Speech analysis method and speech encoding method and apparatus thereof | |
| CN1677490A (en) | Intensified audio-frequency coding-decoding device and method | |
| CN1871501A (en) | Spectrum coding apparatus, spectrum decoding apparatus, acoustic signal transmission apparatus, acoustic signal reception apparatus and methods thereof | |
| CN1875402A (en) | Audio signal encoding or decoding | |
| CN1677493A (en) | Intensified audio-frequency coding-decoding device and method | |
| CN1816847A (en) | Fidelity-optimised variable frame length encoding | |
| CN1161750C (en) | Speech encoding and decoding method and apparatus, telphone set, tone changing method and medium | |
| CN1156872A (en) | Speech encoding method and apparatus | |
| CN1155725A (en) | Speech encoding method and apparatus | |
| CN1527995A (en) | Encoding device and decoding device | |
| CN1702974A (en) | Method and apparatus for encoding/decoding a digital signal | |
| US7844451B2 (en) | Spectrum coding/decoding apparatus and method for reducing distortion of two band spectrums | |
| CN1477872A (en) | Compressed encoding and decoding equipment of multiple sound channel digital voice-frequency signal and its method | |
| AU2012297805A1 (en) | Encoding device and method, decoding device and method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: MATSUSHITA ELECTRIC (AMERICA) INTELLECTUAL PROPERT Free format text: FORMER OWNER: MATSUSHITA ELECTRIC INDUSTRIAL CO, LTD. Effective date: 20140716 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20140716 Address after: California, USA Patentee after: PANASONIC INTELLECTUAL PROPERTY CORPORATION OF AMERICA Address before: Osaka Japan Patentee before: Matsushita Electric Industrial Co.,Ltd. |